Bomb-produced radiocarbon in the shell of the chambered nautilus: rate of growth and age at maturity

Abstract The temporal pattern of bomb-produced radiocarbon in the oceans provides a chronometer to estimate the rate of growth of marine organisms. Radiocarbon measurements by tandem accelerator mass spectrometry of samples from the septa and shell wall of a mature specimen of Nautilus macromphalus reveal progressively increasing values in more recently formed shell material. This pattern is similar to that recorded in a banded coral from surface waters of the Great Barrier Reef, Australia. However, the pre-bomb values of Δ 14 C are lower in the nautilus than in the coral. This offset is probably due to the difference in the respective habitat depths of the two animals. Taking this difference into account, we have used the temporal pattern of Δ 14 C in the coral as a point of reference to help interpret the pattern of Δ 14 C in the nautilus. According to our calculations, the age of the specimen of N. macromphalus is 10–12 years and the timing of septal formation ranges from ⩾ 80 to 240 days/septum.


Introduction
The chambered nautilus is the only surviving member of a large group of extinct shelled cephalopods, and therefore, information about its life history is particularly important. Yet its rate of growth and age at maturity are poorly known. Previous estimates of these parameters have been based on mark-recapture studies [1,2] and observations of Nautilus in aquaria [3][4][5]. An alternative approach involves the use of chronometers incorporated into the shell during growth. Naturally occurring radionuclides of the uranium and thorium decay series have been used to determine the growth rates of a variety of marine organisms [6][7][8][9][10][11][12][13][14][15][16]. In particular, the radionuclides 228Ra (half-life=5.75 y) and 21°Pb (half-life=22 y) present in sea water are incorporated into the shells of marine organisms where they undergo radioactive decay to 228Th (half-life = 1.9 y) and 2a°Po (half-fife= 138 days), respectively. These daughter radionuclides are not directly incorporated into the shells from sea water, but only appear as a result of radioactive decay. Therefore, 0012 the change in the daughter/parent activity ratio with time serves as a chronometer for growth. In Nautilus, 21°po is excluded from the aragonitic shell during its formation, and the 2a°po/Zl°Pb activity ratio increases from 0 to an equilibrium value (1.0) in approximately five half-lives of 2t°po. The 2a°po/Zt°pb activity ratio thus can be used as a chronometer to measure the rate of growth over the two most recent years of growth [15,16]. In a mature specimen, however, the relatively short half-life of 21°po prevents this method from providing information on the age of the animal or its rate of growth early in its ontogeny. The atmospheric testing of thermonuclear weapons during the 1950's and early 1960's introduced fallout radionuclides into the oceans, and the temporal pattern of these tracers also may be used as a chronometer to measure the rate of growth of marine organisms. Bomb-produced radiocarbon increased the A14C (per mil deviation with respect to the activity in 19th century wood) of the atmosphere to nearly 1000%o by 1964. In the surface ocean, the record of A14C is well documented in hermatypic corals [17][18][19][20][21] and ex-hibits an increase above the pre-bomb level starting in the late 1950's and continuing into the mid-1970's. This pattern of a4C variation with time has been used as a chronometer to determine the growth rate of the marine bivalve, Arctica islandica. Turekian et al. [14] used the record of N4C observed in the growth increments of this bivalve to evaluate the hypothesis that these increments form annually. In this paper, we use the record of A14C in the septa and shell wall of a mature specimen of Nautilus macromphalus to infer its age and rate of growth.

Experimental methods and results
The specimen of Nautilus macromphalus was collected in July, 1969 by Dr. J. Redmond (Iowa State University) near Noumea, New Caledonia (22°18'S, 166°48'E). It was caught in traps set between 50 and 80 m depth, although, in fact, this species may live as deep as 600 m [22]. The specimen is 16.5 cm in diameter with 31 septa. The most recent septum is approximated (closely spaced) and thicker than the preceding septum. The angular length of the body chamber, measured from the last septum to the apertural edge, equals 120 °. The approximated last septum and large body chamber angle suggest that this animal is nearly to fully mature [1,23,24].
Six samples of the septa and shell wall were collected for radiocarbon analysis (Fig. 1). Five of the samples are septa ranging from the most recently formed septum (#31) to septa 8 and 9, which were combined to form a single sample. Stable isotopic and morphologic data on other specimens of Nautilus have indicated that septa 8 and 9 are the initial postembryonic septa [15,25]. The sixth sample was obtained by breaking off 3-5 mm of shell from the apertural edge. All samples were rinsed in distilled water to remove any surface material. Sample masses (see Table 1) f!.~~ NUMBER OFSEPTA =51 BODY CHAMBER ANGLE:I20 °  ranged from less than 0.1 g (septa 8/9) to several grams (septum 31). Due to the small masses, particularly for the early septa, 14C was determined by tandem accelerator mass spectrometry (TAMS) at the NSF-Arizona Accelerator Facility for Radioisotope Analysis [26]. Samples were converted to graphite for TAMS and the analytical procedures are detailed elsewhere [27].
To provide a comparison with the N4C record in the nautilus, we present the N4C record in a hermatypic coral from the Great Barrier Reef, Australia. Cores of Porites australiensus were collected from 10 m depth on the outer reef area of Heron Island in November 1983. The coral was sectioned into annual bands and the post-1950 portion measured for radiocarbon according to methods reported by Griffin and Druffel [28]. The radiocarbon measurements for both the nautilus and the coral are reported as N4C according to the definition of Stuiver and Polach [29]. In the nautilus, the A14C values were calculated using ~13C values of the graphite prepared from the carbonate samples.
The results of the a4C measurements of the nautilus samples are listed in Table 1 and the coral data are plotted in Fig. 2. The coral data vary over time from an average of -46%o prior to 1958 to a maximum of 150%o in 1976. In comparison, the nautilus values range from -60 to -71%o in early septa (8/9 and 15) to +68%o in the apertural sample.

Discussion
In constructing a chronology from the nautilus data, we begin by a consideration of the temporal variation in A14C in the surface ocean. Previous work has demonstrated that such variation is recorded in the radiocarbon content of the annual growth bands of corals. Data are available from corals collected in many locations including Bermuda [17], Florida and Belize [18], the Galapagos Islands [20], Tahiti [21], the Fiji Islands [30] and Australia (Fig. 2). The patterns from all these locations are similar in showing an increase in A]4C in surface water starting in the late 1950's and continuing until the 1970's, when the pattern levels off and begins to decrease in temperate regions and continues to increase in tropical regions. Several features of the coral data are relevant to any interpretation of the a4C record in the nautilus. First, the earliest evidence of bomb-produced radiocarbon observed in the surface ocean occurs in 1957-58. Second, corals from different locations exhibit slightly different pre-bomb values of A14C as well as slightly different temporal patterns. These discrepancies are caused chiefly by upwelling which brings deeper water with a lower At4C signature to the surface [17,31]. Corals growing in areas where upwelling occurs on a more or less continuous basis, such as those from the Galapagos Islands, display pre-bomb values of A14C approximately 20%0 lower than those in corals from other areas [20]. In areas in which upwelling occurs sporadically, for example, Bermuda, the values of Aa4C recorded in corals are depressed for the period of time over which the upwelling occurs [17]. The record of N4C in the Australian coral illustrated in Fig. 2 is similar to those from Belize and Florida [18], where upwelling is uncommon. The data from the Australian coral represent a typical temperate ocean trend and we will use this pattern to interpret the pattern in the nautilus.
The pattern of N4C in the nautilus is similar to that in the Australian coral. In the nautilus the values of N4C in septa 8/9 and 15 are low and approximately the same within the reported one sigma error (-60_+ 12%o and -71 _+ 12%o). The rest of the samples exhibit a monotonic increase to positive values of A14C reaching a maximum of +68 _+ 6%o in the apertural sample. Based on comparison with the coral record, we interpret septa 8/9 and 15 as having formed prior to the first appearance of bomb radiocarbon ifi the water in which the nautilus grew its shell. In the coral, the pre-bomb values of N4C average -46%o. In the nautilus the average pre-bomb values are lower than those in the coral, --66%o. This difference may be due to upwelling as described above, but it is more likely due to the fact that N. macromphalus is a mobile animal and, in New Caledonia, commonly lives at depths between 300 and 500 m [22]. Thus, on the average, the nautilus is living in deeper water than the coral and this difference in depth is reflected in its record of A14C. Broecker and Peng [31], using water column data collected in the 1950's, reconstructed the pre-bomb depth profile of A]4C in the South Pacific. Values range from approximately -50%o in the surface ocean (comparable to that recorded in the Australian coral prior to 1959) to --100%o at a depth of 800 m. A pre-bomb AI4c value of -66%0 occurs at depths of about 250-350 m, well within the range of Nautilus. It thus appears that the nautilus is recording the radiocarbon content at depth in the water column, and therefore, its record of A14C will probably be offset from that of the Australian coral much like the radiocarbon record of the Galapagos coral is offset from those of other temperate corals.
In order to assign a chronology to the record of A~4C in the nautilus, we assume that the regular increase in At4C from the oldest to most recently formed shell material results from shell secretion at a consistent depth or range of depths. We also assume that: (1) septa 8/9 and 15 were formed prior to the first appearance of bomb radiocarbon in the water in which the nautilus grew and the average value of A14C in these samples represents the baseline, pre-bomb level, (2) septa 20, 25, and 31 were formed after 1958 (the last year before the appearance of bomb radiocarbon in the Australian coral) but before 1969, the date of collection of the nautilus specimen, and (3) the apertural sample was formed in 1969. Our task is then to reconstruct the temporal change in A]4C in the water in which the nautilus grew and compare this change with that recorded in the nautilus shell.
We have information on the depth profiles of A14C at two times in the South Pacific: pre-bomb, as mentioned above (1955-1957, [31 (Fig. 3a). The large filled circles represent the nautilus A]4C data.

31
values expressed in the nautilus data. At 200 m, the values of N4C range from -57%o (pre-bomb) to +68%o at 1969 and are plotted in Fig. 3b. Between 0 and 360 m, the depth-weighted average values of N4C span about the same range, -57%o to + 69%o, and are plotted in Fig. 3c. In terms of  the habitat of the nautilus, Fig. 3b represents the temporal pattern of A14C observed by the animal We have assigned dates to the values of A14C in the nautilus samples on the basis of Fig. 3b and 3c and the results are listed in Table 2. Although the two methods yield similar results as the bomb radiocarbon signal penetrates deeper into the water column over time, the results differ for the early record. The method of averaging from 0 to 360 m implies that the bomb radiocarbon signal is incorporated into the shell shortly after its first appearance in the surface water (1958), whereas the 200 m method delays the first appearance of bomb radiocarbon in the shell until this signal has penetrated to a depth of 200 m (-1960-61). In both approaches the age of the aperture is approximately 3 years older than the most recent septum (#31), suggesting that the animal continued growing at the aperture after secreting the last septum. This result is supported by the large angular length of the body chamber, which is characteristic of mature specimens [1,23,24].
Using the ~14C chronologies, we calculated the time intervals required for septal formation. The values range from 130 to 240 days/septum for septa 20 to 31. Within the uncertainties of the measurements listed in Table 2, these values compare well with previous estimates of 50-100 days/septum in N. macromphalus maintained in aquaria [5] and up to 300 days/septum in N.
The methods we have used to assign ages to the nautilus samples do not allow us to calculate the dates at which septa 8/9 and 15 formed because these septa predated the first appearance of the bomb 14C signal. However, we can set lower limits on the ages of these samples. The bomb 14C signal was first recorded in the nautilus shell between septa 15 and 20. If we assign the last year before the bomb 14C signal first appeared in the water in which the nautilus grew as the date of formation of septum 15 (Fig. 3b or 3c), we produce a lower limit for the time elapsed between septa 15 and 20 (and a lower limit for the timing of septal formation in this interval). The assigned dates depend on the method used to construct the •4C vs. time a Both methods are based on reconstruction of AI4c depth profiles over time (Fig. 3). Method I uses the temporal variation in Alac values at a depth of 200 m whereas method II uses a weighted average of A14C values in the upper 360 m of the water column. In both approaches the A14C values range from --57%0 to +68-69%0, comparable to those observed in the specimen of N.

macromphalus.
b Assumes that septum 15 was the last septum formed before the first appearance of bomb radiocarbon.  Table 2). The corresponding times of septal formation are >~ 150 days/septum or >~ 80 days/septum respectively ( Table 2). Although these values are lower limits, they are comparable to the estimates cited earlier based on other studies.
Finally, we would like to use the 14C chronology to calculate the age of the specimen and hence its age at maturity. Our samples bracket the entire postembryonic development of the animal because septa 8/9 are the first septa formed after the animal hatched from its egg capsule [15,25,33]. In addition to our previous calculations, we must calculate the time elapsed between septa 8/9 and 15 in order to calculate this age. We adopt a conservative approach and use the shortest time of septal formation reported for N. macromphalus, -50 days/septum. This value, based on aquarium studies [5], yields an elapsed time of approximately one year between septa 8/9 and 15. The age at maturity of this specimen of N. macromphalus is, therefore, at least 10-12 years, depending on the method used to construct the 14C chronology (Table 2). This estimate exceeds values of 2.5 to 6 years for this species based on aquarium studies [4] but is comparable to values of 10-15 years for N. belauensis in nature based on markrecapture and radiometric studies [1,16]. These results emphasize the striking difference between the life history of Nautilus and that of most other modern cephalopods, which rapidly reach maturity in 1-3 years [34].